E. coli K1 is a leading cause of septicemia and meningitis in newborn infants. Conventional antibiotic therapy is often ineffective: A better understanding of the genetic basis for K1 colonization and invasion might suggest new modes of prophylaxis or treatment. Little is known about how K1 colonizes and causes invasive disease, or about the gene regulation permitting transition from mucosal surface colonization to growth and survival in normally sterile host compartments. However, it is known that the closely related laboratory strain, E. coli K-12, is unable to persist in either the intestinal tract or the bloodstream. Thus information on the genetic determinants of colonization and invasion is contained in the differences between the largely homologous genomes of these two bacteria. In this project, the determinants of pathogenesis in E. coli K1 are mapped by replacing segments of its chromosome, one at a time, with the corresponding K-12 sequences. This approach of 'chromosome replacement', or E. coli K-12 map-directed 'mutagenesis', employs a set of bacteriophage P1-linked transposons - of alternating antibiotic resistance - that function ally divide the entire chromosome of wild-type E. coli K-1 2 (strain MG1 655) into 107 distinct P1-trans ducible intervals (Singer, et al., Micro. Rev. 1989). Segments of the K1 chromosome are replaced in two steps: First, a transposon Tn 10kan flanking one end of the targeted segment is transduced into the prototype rat-virulent E. coli K1 strain, RS218. Second, transduction of the P1-linked Tn 10 bordering the other end of the segment - into this RS218::Tnl0kan - is selected on tetracycline. Linkage of the two markers, indicated by transduction to kanamycin sensitivity in the second step, occurs by replacement of the intervening region of the K1 chromosome with K-12 DNA. The Kl/K-12 hybrids are screened for loss of colonization and/or invasion in neonatal rats. Nonpathogenic hybrids will be used (i) to probe a K1-minus-K-12 genomic subtraction library (Straus, Church, and Ausubel; in preparation) for K1 virulence factors; (ii) to clone both allelic and nonallelic K1 virulence factors by complementation, and (iii) to detect - by mutagenesis of the substituted K-12 sequence - possible K-12 inhibitors of pathogenesis. The unaffected hybrids will constitute a library of K-12 DNA in a oolonization-proficient/pathogenic background: By mutational analysis of the substituted K-12 sequences, cryptic regions of the K-12 genome (i.e., cryptic in vitro) can be assessed for in vivo phenotypes. Thus the primary goal,of the project - to map and analyze the determinants of K1 pathogenesis - will also allow the genetic tools, sequence data, gene-protein index data, and wealth of previous experimentation in K-12 to be applied to the biology of E. coli in the context of the mammalian host.